Trophic Assessment in Chinese Coastal Systems - Review of Methods and Application to the Changjiang (Yangtze) Estuary and Jiaozhou Bay YONGJIN XIAO 1 ,JOA ˜ O G. FERREIRA 1 ,SUZANNE B. BRICKER 2, *, JOA ˜ O P. NUNES 1 ,MINGYUAN ZHU 3 , and XUELEI ZHANG 3 1 IMAR – Institute of Marine Research, Center for Ocean and Environment, Departmento Cie ˆncias e Eng. Ambiente, Fac. Cie ˆncias e Tecnologia, Qta Torre, 2829-516 Monte de Caparica, Portugal 2 National Oceanic and Atmospheric Administration, National Ocean Service, National Center for Coastal Ocean Science, 1305 East West Highway, Silver Spring, Maryland 20910 3 Research Center for Marine Ecology, First Institute of Oceanography, State Oceanic Administration, Qingdao 266061, China ABSTRACT: Coastal eutrophication has become one of the main threats to Chinese coastal areas during the last two decades. High nutrient loads from human activities have modified the natural background water quality in coastal water bodies, resulting in a range of undesirable effects. There is a need to assess the eutrophic level in coastal systems and to identify the extent of this impact to guide development of appropriate management efforts. Traditional Chinese assessment methods are discussed and compared with other currently-used methods, such as the Oslo-Paris Convention for the Protection of the North Sea (OSPAR) Comprehensive Procedure and Assessment of Estuarine Trophic Status (ASSETS). The ASSETS method and two Chinese methods were tested on two Chinese systems: the Changjiang (Yangtze) Estuary and Jiaozhou Bay. ASSETS is process based, and uses a pressure-state-response model based on three main indices: Influencing Factors, Overall Eutrophic Condition, and Future Outlook. The traditional methods are based on a nutrient index. ASSETS was successfully applied to both systems, classifying the Changjiang Estuary as Bad (high eutrophication) and Jiaozhou Bay as High (low eutrophication). The traditional methods led to ambiguous results, particularly for Jiaozhou Bay, due to the high spatial variability of data and a failure to assess the role of shellfish aquaculture in nutrient control. An overview of the Chinese coastal zone identifies 50 estuaries and bays that should form part of a national assessment. A comparison of methods and results suggests that ASSETS is a promising tool for evaluating the eutrophication status of these systems. Introduction Eutrophication of the coastal zone has received increasing scientific attention worldwide, resulting in the publication of about 5,000 journal articles over the last two decades (SCIRUS 2007). Although China is subject to a huge human-induced nutrient modification in coastal systems, only about 10% of these articles address this region, and very few (e.g., Wang 2006) deal with eutrophication assessment. The strong development of the Chinese econo- my, centered mainly on manufacturing, together with the influx of rural populations to urban areas, many of which are located in the coastal zone or near major rivers, have resulted in a substantial increase in nutrient loads, leading to the prolifer- ation of phytoplanktonic blooms (Guo et al. 1998; Hao et al. 2000; Shen 2001). Frequent occurrences of harmful algal blooms (HAB; Fig. 1) and other eutrophication symptoms have become serious issues in Chinese coastal systems (Harrison et al. 1990; Zhang et al. 1999a; Huang et al. 2003). In 2006, a total of 93 red tide events were observed in Chinese national marine waters, affecting an area of about 20,000 km 2 (P. R. C. State Oceanic Adminis- tration 2007). Although it can be argued that the increase in recorded HAB incidents is partly due to improvements in the national monitoring network, these figures clearly indicate the increasing devel- opment of serious eutrophication problems on a national scale. The potential ecological consequences of nutri- ent enrichment, such as loss or degradation of seagrass beds, interdiction of shellfish aquaculture, and fish kills, are well established (Stevenson et al. 1993; Burkholder et al. 1999; Tomasko et al. 2001; Hauxwell et al. 2003; Wazniak and Glibert 2004) and strongly shape public concern and scientific research for better understanding of eutrophication (e.g., Cloern 2001; Tett et al. 2003, 2007). In order to set priorities for managing and mitigating nutrient enrichment, there is a need for the nation to perform an assessment to * Corresponding author; tele: 301/713-3020 ext. 139; fax: 301/ 713-4388; e-mail: [email protected] Estuaries and Coasts Vol. 30, No. 6, p. 901–918 December 2007 ß 2007 Estuarine Research Federation 901
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Trophic Assessment in Chinese Coastal Systems - Review of
Methods and Application to the Changjiang (Yangtze) Estuary
and Jiaozhou Bay
YONGJIN XIAO1, JOAO G. FERREIRA1, SUZANNE B. BRICKER2,*, JOAO P. NUNES1, MINGYUAN ZHU3, andXUELEI ZHANG3
1 IMAR – Institute of Marine Research, Center for Ocean and Environment, Departmento Ciencias e Eng.Ambiente, Fac. Ciencias e Tecnologia, Qta Torre, 2829-516 Monte de Caparica, Portugal
2 National Oceanic and Atmospheric Administration, National Ocean Service, National Center forCoastal Ocean Science, 1305 East West Highway, Silver Spring, Maryland 20910
3 Research Center for Marine Ecology, First Institute of Oceanography, State Oceanic Administration,Qingdao 266061, China
ABSTRACT: Coastal eutrophication has become one of the main threats to Chinese coastal areas during the last twodecades. High nutrient loads from human activities have modified the natural background water quality in coastal waterbodies, resulting in a range of undesirable effects. There is a need to assess the eutrophic level in coastal systems and toidentify the extent of this impact to guide development of appropriate management efforts. Traditional Chinese assessmentmethods are discussed and compared with other currently-used methods, such as the Oslo-Paris Convention for theProtection of the North Sea (OSPAR) Comprehensive Procedure and Assessment of Estuarine Trophic Status (ASSETS). TheASSETS method and two Chinese methods were tested on two Chinese systems: the Changjiang (Yangtze) Estuary andJiaozhou Bay. ASSETS is process based, and uses a pressure-state-response model based on three main indices: InfluencingFactors, Overall Eutrophic Condition, and Future Outlook. The traditional methods are based on a nutrient index. ASSETSwas successfully applied to both systems, classifying the Changjiang Estuary as Bad (high eutrophication) and Jiaozhou Bay asHigh (low eutrophication). The traditional methods led to ambiguous results, particularly for Jiaozhou Bay, due to the highspatial variability of data and a failure to assess the role of shellfish aquaculture in nutrient control. An overview of theChinese coastal zone identifies 50 estuaries and bays that should form part of a national assessment. A comparison ofmethods and results suggests that ASSETS is a promising tool for evaluating the eutrophication status of these systems.
Introduction
Eutrophication of the coastal zone has receivedincreasing scientific attention worldwide, resultingin the publication of about 5,000 journal articlesover the last two decades (SCIRUS 2007). AlthoughChina is subject to a huge human-induced nutrientmodification in coastal systems, only about 10% ofthese articles address this region, and very few (e.g.,Wang 2006) deal with eutrophication assessment.
The strong development of the Chinese econo-my, centered mainly on manufacturing, togetherwith the influx of rural populations to urban areas,many of which are located in the coastal zone ornear major rivers, have resulted in a substantialincrease in nutrient loads, leading to the prolifer-ation of phytoplanktonic blooms (Guo et al. 1998;Hao et al. 2000; Shen 2001). Frequent occurrencesof harmful algal blooms (HAB; Fig. 1) and othereutrophication symptoms have become serious
issues in Chinese coastal systems (Harrison et al.1990; Zhang et al. 1999a; Huang et al. 2003). In2006, a total of 93 red tide events were observed inChinese national marine waters, affecting an area ofabout 20,000 km2 (P. R. C. State Oceanic Adminis-tration 2007). Although it can be argued that theincrease in recorded HAB incidents is partly due toimprovements in the national monitoring network,these figures clearly indicate the increasing devel-opment of serious eutrophication problems on anational scale.
The potential ecological consequences of nutri-ent enrichment, such as loss or degradation ofseagrass beds, interdiction of shellfish aquaculture,and fish kills, are well established (Stevenson et al.1993; Burkholder et al. 1999; Tomasko et al. 2001;Hauxwell et al. 2003; Wazniak and Glibert 2004)and strongly shape public concern and scientificresearch for better understanding of eutrophication(e.g., Cloern 2001; Tett et al. 2003, 2007).
In order to set priorities for managing andmitigating nutrient enrichment, there is a needfor the nation to perform an assessment to
* Corresponding author; tele: 301/713-3020 ext. 139; fax: 301/713-4388; e-mail: [email protected]
Estuaries and Coasts Vol. 30, No. 6, p. 901–918 December 2007
� 2007 Estuarine Research Federation 901
determine the nature and scope of these problems.A range of eutrophication assessment proceduresare currently applied in Chinese waters, such asNutrient Index Methods. These approaches arederived from freshwater methods and are notnecessarily appropriate to coastal systems, since itis commonly accepted that coastal eutrophication isa far more complex problem, as pointed out by Mayet al. (2003); Nunes et al. (2003), and Ferreira et al.(2005). These methods remain at the Phase I stage(sensu Cloern 2001) in the development of eutro-phication assessment methods, focusing on nutrientconcentrations rather than on complex direct andindirect ecosystem responses.
Although efforts have been made to improvereliability and precision, e.g. through the applica-tion of Principal Component Cluster Analysis (PCA;Lin et al. 2004) and Fuzzy Analysis (Xiong and Chen1993), these mainly aim to optimize the applicationof Phase I methods, but fail to develop a newrationale for the assessment methodology. With theadvances in our understanding of eutrophication,Phase II methods, defined as symptoms-based andmulti-indicator, have the merit of taking intoaccount system responses, and for some methods,pressures as well. Well known examples include theUnited States Environmental Protection Agency(USEPA) National Coastal Assessment (NCA) WaterQuality Index method (USEPA 2005), the NationalOceanic and Atmospheric Administration’s Nation-al Estuarine Eutrophication Assessment (NEEA;Bricker et al. 1999) and Assessment of EstuarineTrophic Status (ASSETS; Bricker et al. 2003), andthe Oslo-Paris Convention for the Protection of theNorth Sea Comprehensive Procedure (OSPARCOMPP; OSPAR Commission 2003).
These Phase II methods develop well-establishedscientific theory and have been successfully appliedin America and Europe; none have been tested inChinese coastal systems. This paper aims to providean overview of Chinese coastal systems in order to
define the spatial extent of the management issue,and to compare and contrast the Chinese Phase Imethods with more recent approaches to helpinform the choice of methods for integratedassessment.
Two Chinese systems, the Yangtze Kiang (Chang-jiang or Long River) Estuary, near Shanghai, andJiaozhou Bay, near Qingdao, were chosen as testsites for this work. The contrasting nature of the twosystems is appropriate for evaluating the widerapplication of these eutrophication assessmentmethods in China. As the Chinese coast encom-passes a large area and about 20 degrees of latitudeand longitude, this evaluation potentially contrib-utes to improved coastal management in othersoutheast Asian nations.
REVIEW OF CHINESE COASTAL SYSTEMS
The Chinese coastal zone covers 23u of latitude(17uN to 40uN) and 16u of longitude (108uE to124.5uE) and has an area of 2.85 3 105 km2. Thesecoastal areas are in general densely populated andsubject to intense economic activity: the waterbodies are often characterized by important anthro-pogenic nutrient loads.
As a result of these pressures, eutrophication isone of the most negative factors influencingecosystem health of Chinese coastal systems. Coastalareas of all the major Chinese seas (Bohai,Huanghai, Donghai, and Nanhai) are a concernwith respect to HAB (Fig. 1). HAB occurrence wasfirst documented in the 1930s, and since then, thenumber and scale of HAB events appear to beincreasing over time (Yan et al. 2002).
Physical and water quality data for Chinese baysand estuaries were collated based on Chinesereference material (Editorial Board of Bays in China1993, 1998) in order to provide an overview of thespatial scope of the management issue. These dataare largely from the 1980s, and have been condensedin Table 1 to provide a synthesis for 16 major coastalsystems representing 95% of the overall area.
Although a national trophic assessment of Chi-nese coastal systems has not yet been conducted, itis known from various sources, both scientific andanecdotal, that most coastal areas and estuariesappear to exhibit nutrient-related eutrophicationsymptoms. The literature suggests that those systemsundergoing most severe eutrophication include theBohai Bay, Changjiang Estuary, Hangzhou Bay, andPearl River Estuary (Zou et al. 1985; Peng and Wang1991; Pei and Ma 2002; Chai et al. 2006).
ASSESSMENT METHODS
Eutrophication assessment began with the classi-cal freshwater approach (Vollenweider 1968, 1975;
Fig. 1. Harmful algal bloom events in coastal China from 1972to 2004 and regional proportion in the last decade (http://www.china-hab.cn).
902 Y. Xiao et al.
Carlson 1977; Morihiro et al. 1981) and hasdeveloped through two phases as discussed earlier(Cloern 2001). Historically, a number of eutrophi-cation assessment methods have been proposed inChina, such as the Nutrient Index Method, PCA,and Fuzzy Analysis. These methods focus on theevaluation, using chemical tools, of effects of systemloading by nutrients, and may be considered Phase Iapproaches (Wang 2005; Yao and Shen 2005).These methods are reviewed in this section,together with three Phase II methods: OSPAR-COMPP, NCA Water Quality Index, and ASSETS.These more current approaches will potentially helpthe evaluation of eutrophication in Chinese coastalwaters.
NUTRIENT INDEX METHOD I
This method, proposed by the Chinese NationalEnvironmental Monitoring Center, is based on anutrient index (NI) in seawater (Lin 1996), calcu-lated by using Eq. 1:
Ni ~CCOD
SCODz
CTN
STNz
CTP
STPz
CChla
SChlað1Þ
where: CCOD, CTN, CTP, and CChla are measuredconcentrations of chemical oxygen demand (COD),total nitrogen, total phosphorus (all in mg l21), andchlorophyll a (in mg l21) in seawater, respectively.SCOD, STN, STP, and SChla are standard concentrationsof COD (3.0 mg l21), total nitrogen (0.6 mg l21),total phosphorus (0.03 mg l21), and chl a(10 mg l21) in seawater, respectively (Lin 1996). IfNi is greater than four the seawater is consideredeutrophic.
NUTRIENT INDEX METHOD II
Zou et al. (1985) proposed an alternative nutrientindex method, adapted from Japanese assessmentmethods (Okaichi 2004):
Ni ~CCODCDIN CDIP
Scð2Þ
where CCOD, CDIN, and CDIP are measured concen-trations (in mg l21) of COD, dissolved inorganicnitrogen, and dissolved inorganic phosphorus inseawater, respectively. Sc in Eq. 2 is the meanproduct of standard concentrations of COD, DIN,and DIP, for which a constant value of 4.5 3 1023 isused as it is believed that the critical value for CODis 1–3 mg l21, DIN is 0.2–0.3 mg l21, and DIP is0.01–0.02 mg l21 (Chen et al. 2002). If Ni, is greaterthan one the seawater is considered eutrophic.
These two limnology-originated methods aresimple to apply and use indicators that are easy todetermine. While they are widely used in Chinesecoastal systems, research in the past decades hasidentified key differences in the responses of lakesand coastal-estuarine ecosystems to nutrient enrich-ment (Cloern 2001; Bricker et al. 2003; Ferreira etal. 2007a), and the adaptation of approaches usedfor freshwater has often met with limited success incoastal areas. This is partly because in coastalenvironments there is often no clear relationshipbetween nutrient forcing and eutrophication symp-toms—systems with similar pressures show widelyvarying responses. Nutrient concentrations haveoften been shown to be poor indicators of eutro-phication symptoms (e.g., Tett et al. 2003), sinceecosystem responses are modulated by typological
TABLE 1. Overview of Chinese coastal systems condensed from a set of 50 major coastal systems, with size ranges that are consideredmedium (, 400 km2), large (400–650 km2), and extra large (. 650 km2). Data from Jiaozhou Bay (excluding area) are median values of150 water quality samples at seven stations, collected in 1999–2000 by the European Union International Cooperation for DevelopingCountries (INCO-DC) project: Carrying Capacity and Impact of Aquaculture on the Environment in Chinese Bays.
System Area (km2) NH4 (mmol l21) NO2 (mmol l21) NO3 (mmol l21) PO4 (mmol l21) N:P Chl a (mg l21)
Changjiang Estuary 51,000 1.40–20.0 0.10–2.50 68.0 20.7 3.40–3.50 1.13Hangzhou Bay 5,000 9.86 1.71 112 1.13 110 NALeizhou Bay 1,690 NA 0.39 2.34 0.13 NA NAWenzhou Bay 1,474 1.41 0.39 14.2 1.17 13.7 1.54Honghai Bay 925 2.50 0.23 5.30 0.24 33.5 3.12Taizhou Bay 912 NA 0.40 27.3 0.65 NA 2.15Haizhou Bay 876 0.75 0.10 1.50 0.11 21.4 NASanmen Bay 775 1.36 0.36 16.1 0.94 19.0 1.47Sansha Bay 570 1.10 1.02 11.2 0.66 20.2 0.79Pulandian Bay 530 NA 0.29 2.20 0.26 NA 5.68Daya Bay 516 0.21 0.20 0.56 0.20 4.90 1.70Zhanjiang Gang 490 NA 0.64 9.29 0.14 NA NAYueqing Bay 464 0.80 0.27 31.1 0.72 44.7 1.40Meizhou Bay 424 1.14 0.75 7.20 0.33 27.5 1.70Aiwan-Xuanmen Bay 419 1.25 0.48 14.9 0.72 23.1 1.00Jiaozhou Bay 397 4.62 0.64 3.77 0.26 33.5 1.67
Trophic Assessment in Chinese Coastal Systems 903
factors, such as morphology, tidal range, naturalturbidity, and water residence time.
PRINCIPAL COMPONENT CLUSTER ANALYSIS
PCA is applied to eutrophication assessmenton the premise that traditionally sampled indi-cators are correlated to each other, enabling theprincipal components to be obtained from theset of indicators. Two main types of trophicindicators have been (and are still) used ineutrophication studies: physico-chemical and bio-logical factors (Parinet et al. 2004). Althoughthese factors are clearly related (Strain and Yeats1999), the form of the relationship may varyconsiderably. Given the complexity of coastalsystems, the aim of this method is to apply linearregression to provide a more reliable way tocharacterize the state of the aquatic system throughan aggregation approach rather than throughindividual indicators.
The application of PCA yields a subset ofindicators that allow a simplified assessment of asystem, while retaining the key information. Aftercalibration, the selection of principal indicators maybe extended to other systems, with the caveat thatdifferences in typology may render the subsetinapplicable.
Since most coastal systems have data gaps, PCApotentially allows scientists or managers to makebetter use of available information, due to thereduced subset of indicators required. In combina-tion with Nutrient Index Methods, PCA is able toidentify the most important indicators related toeutrophic conditions, providing more flexibilitythan the use of indicators from the Nutrient IndexMethods themselves. Because PCA is based purelyon statistical rationality, and since so few parametersare used in Nutrient Index Methods, it has not beencommonly applied to Chinese systems.
FUZZY ANALYSIS
The main advantage of Fuzzy Analysis is the abilityto deal with imprecise, uncertain, or ambiguousdata or relationships, which clearly fits the study ofecological and environmental issues (Metternicht2001). In most conventional methods, a variety ofthreshold values are used to give a classification forindicators when evaluating the system status. Adiscrepancy frequently arises from the lack of a cleardistinction between the uncertainty in the qualitycriteria employed and the vagueness or fuzzinessembedded in the decision-making output values(Chang et al. 2001). Owing to inherent imprecision,difficulties always exist in describing eutrophicconditions through distinct numbers used asthresholds for various indicators.
Early eutrophication index methods, such asCarlson’s index, tended to categorize the eutrophi-cation level into discrete numbers (Carlson 1977).Trophic State Index values of 49 and 50 are indifferent classes while 41 falls into the same categorywith 49, even though it might be more reasonable toput 49 and 50 together. In this case, Fuzzy Analysisseems to be a possible solution to deal with theambiguity within eutrophication assessment. Al-though the theory had existed for decades, theapplication of Fuzzy Analysis to assess water qualitybegan in the 1990s (Peng and Wang 1991; Kung etal. 1992; Salski 1992; Lu and Lo 2002; Marchini andMarchini 2006).
Despite substantial efforts in development ofFuzzy Analysis based on ecological models (Kom-pare et al. 1994; Chen and Mynett 2003b), progresshas been slow for two reasons: highly dimensionalsystems require a large and redundant ruleset, thesize of which grows exponentially with the numberof indicators; and membership functions andinference rules are difficult to define.
OSPAR COMPP
COMPP has been adopted by OSPAR for theidentification of eutrophication status of the OSPARMaritime Area (OSPAR 2003). OSPAR COMPP is astepwise method that consists of two main proce-dures: the Screening Procedure and the Compre-hensive Procedure. The Screening Procedure is abroad-brush approach, designed to identify obviousnon-problem areas with respect to eutrophication.Areas that are not identified as obvious non-problem areas in the first procedure are subjectedto the Comprehensive Procedure and classified intoproblem areas, potential problem areas, and non-problem areas. In OSPAR COMPP the requiredsampling frequency and spatial coverage of allindicators are dependent on the final classificationof the areas.
The first step of OSPAR COMPP is the classifica-tion of assessment indicators to provide a score foreach of the assessment criteria. Category I is scored+ in cases where one or more of its respectiveassessment indicators show an increased trend,elevated change, or elevated concentration (i.e.,greater than the reference value + 50%). Thesecond step is to integrate the scores obtained fromthe first step so as to provide a coherent classifica-tion of the area. An evaluation is made for eachassessment indicator from four categories (Influ-encing Factors, Direct Effects, Indirect Effects,Other Possible Effects) determining whether mea-sured concentrations relate to a problem area,potential problem area, or non-problem area. Thefinal step is to combine the results for the fourcategories, taking into account supporting environ-
904 Y. Xiao et al.
mental factors and region-specific characteristics,such as physical and hydrodynamic aspects andweather-climate conditions, to make a final rating ofproblem, potential problem, or non-problem area.The region-specific characteristics play a role inexplaining the results of the area classification andare essential for the definition of a final classifica-tion.
NCA WATER QUALITY INDEX
To summarize the condition of ecological re-sources in the coastal waters of the United States,the USEPA developed a Water Quality Index,outlined in the National Coastal Condition ReportII (USEPA 2005). The Water Quality Index consistsof five indicators: DIN, DIP, chl a, water clarity, anddissolved oxygen (DO). Each of the indicators isclassified into one of three categories: good, fair, orpoor. The overall water quality index uses theresults for each of the five indicators to calculatean overall rating that also falls into one of thesethree categories, but unlike OSPAR COMPP andASSETS, the Water Quality Index does not includean evaluation of influencing factors.
ASSETS
ASSETS was developed from the NEEA, which wasapplied to 141 estuaries in the U.S. ASSETS is amore sophisticated and integrated method foreutrophication assessment in coastal zones, whichmay be applied comparatively to rank the eutrophi-cation status of estuaries and coastal areas. TheASSETS approach includes quantitative and semi-quantitative components, and uses a combination offield data, models, and expert knowledge toevaluate pressure-state-response indicators. Thecore methodology relies on three diagnostic tools:a simple model to estimate pressure (InfluencingFactors), a symptoms-based evaluation of state(Overall Eutrophic Conditions), and an indicatorof expected future conditions (Future Outlook). Itcombines primary (chl a, macroalgae) and second-ary (DO, nuisance-toxic algal blooms, spatial chang-es of submerged aquatic vegetation [SAV] distribu-tion) symptoms to derive an Overall EutrophicCondition index, which is then associated with ameasure of Influencing Factors and Future Out-look. ASSETS may be divided into three parts: datacollection and compilation, application of indices(Table 2; for a full description see Bricker et al.2003), and grouping and synthesis.
Pressure - Influencing Factors
The approach for Influencing Factors considersthat systems exhibit varying symptoms or symptomlevels as a consequence of a particular nutrient load,
due to differential susceptibility to nutrient inputs(Bricker et al. 1999). System susceptibility is definedas the relative capacity of a system to dilute andflush nutrients, and is determined by systemvolume, tidal range, mixing, and river flow. Nutrientinputs describe the comparison of nutrients fromwatershed or land-based (human) loads with oce-anic or natural loads. Susceptibility and nutrientinputs are combined in a matrix to determine thefinal Influencing Factors rating.
State - Overall Eutrophic Condition
The Overall Eutrophic Condition index uses asequential approach based on two groups ofsymptoms, which bring together five indicators:chl a and macroalgae, which are indicators forprimary symptoms, and loss of SAV, DO, andnuisance-toxic blooms, which indicate secondarysymptoms. The primary symptoms correspond tothe early stage of water quality degradation andpotentially lead to well-developed eutrophic condi-tions, i.e., secondary (advanced) symptoms, such asSAV loss, nuisance-toxic algal blooms, and low DO(anoxia or hypoxia).
The level of expression of the primary symptomsis determined by calculating the average of twoprimary symptom expression values, with the chl aexpression level determined as the 90th percentileof annual data values. The level of secondarysymptoms is obtained in a precautionary mannerby choosing the worst of three symptoms, with DOexpression level determined from 10th percentilevalues of annual data. The primary and secondarysymptoms are combined in a matrix to determinean overall level of eutrophic conditions for theestuary.
Response - Future Outlook
An analysis of Future Outlook is performed todetermine whether conditions in an estuary or baywill worsen, improve, or stay the same in themedium term (e.g., over the next two decades).Assessment of expected changes in nutrient pres-sures is performed based on a variety of drivers,including demographic trends, wastewater treat-ment, and remediation plans, together with expect-ed changes in agricultural practices and watersheduses. Projection of future nutrient inputs is com-bined with system susceptibility to predict futurescenarios.
Synthesis - Overall Grade
The final stage of ASSETS is to synthesize thethree indices mentioned above to provide an overalldescription of system status in terms of eutrophica-tion. The combination of individual classifications
Trophic Assessment in Chinese Coastal Systems 905
for pressure, state, and response is able to provide agrade falling into one of five categories: high(better), good, moderate, poor, or bad (worse).These grades match the quality classes of theEuropean Union 2000/60/EC (Water Framework)Directive (European Community 2000).
DISCUSSION OF METHODS
In terms of indicator variables, the four Chinesemethods outlined above are quite similar althoughthe underlying logic may vary. The indicatorsapplied in the different methods are summarizedin Table 3, and Table 4 presents a more detailedcomparison among Chinese methods and Phase IImethods.
Even though Nutrient Index Methods are theapproach recommended by the Chinese NationalAuthorities, these methods have been criticized fortheir underlying limitations and simplicity (Yao andShen 2005). There are two main reasons that the
Phase I methods are considered inadequate for theassessment of Chinese coastal eutrophication. First,there has been an overemphasis on the significanceof nutrient concentration as an indicator ofeutrophication, which may not be a robust diagnos-tic indicator. Nutrients are the primary cause, butthere are many factors causing or responding to theincrease of eutrophic level, such as the presence ofnuisance algae and loss of SAV. High nutrientconcentrations are not necessarily indicative ofeutrophication, and low concentrations do notunequivocally guarantee the absence of eutrophica-tion (Cloern 2001; Dettmann 2001; Bricker et al.2003). Second, there has been a failure in adapta-tion of a freshwater-based approach to coastal watersdue to the differences between freshwater andcoastal systems. For example, water exchange ortop-down control by filter feeders (of great rele-vance in China due to the intensity of shellfishaquaculture) in estuaries and bays may greatly
TABLE 2. List of indicators considered in ASSETS (adapted from Bricker et al. 2003).
Indicators Existing Conditions
Chlorophyll a Surface concentrations:Limiting factors to algal biomass (N, P, Si, light, other)Spatial coveragea; month of occurrence; frequency of occurrenceb
Nuisance-toxic algae Occurrence: problem (significant effect upon biological resources); no problem (noDominant speciesEvent duration (hours, days, weeks, seasonal, other)Months of occurrence; frequency of occurrenceb
Macroalgae Abundance: problem (significant effect upon biological resources); no problem (nosignificant effect)
Months of occurrence; frequency of occurrenceb
Anoxia (0 mg l21) Dissolved oxygen condition: observed, no occurrenceHypoxia (0–2 mg l21) Stratification (degree of influence): high, medium, low, not a factorBiological stress (2–5 mg l21) Water column depth: surface, bottom, throughout water column
Spatial coveragea; month of occurrence; frequency of occurrenceb
Submerged aquatic vegetation-intertidal wetlands Spatial coverage (loss, gain, no change)
a Spatial coverage (% of salinity zone): high (50–100%), medium (25–50%), low (10–25%), no SAV-wetland in system.b Frequency of occurrence: episodic (conditions occur randomly), periodic (conditions occur annually or predictably), persistent
(conditions occur continually throughout the year).
TABLE 3. Summary of indicator variables used (adapted from Bricker et al. 2006).
IndicatorsNutrientIndex I
NutrientIndex II PCA Fuzzy Analysis OSPAR COMPP EPA NCA ASSETS
Nutrient (DIN, DIP) load or concentration 3 3 3 3 3 3 3
Chemical oxygen demand 3 3Chlorophyll a 3 3 3 3 3 3Dissolved oxygen 3 3 3 3 3 3 3Water clarity 3 3 3HABs/Nuisance 3 3Phytoplankton indicator species 3Macroalgal abundance 3 3
Submerged aquatic vegetation loss 3 3Zoobenthos-fish kills 3
906 Y. Xiao et al.
mitigate the expression of eutrophication symp-toms, even under high nutrient loading. Simpletime-varying or statistical approaches established infreshwater, based on the clear relationships betweenpelagic algae and nutrient loading, contrast withambiguous relationships that may occur in coastalwaters (Ferreira et al. 2007a).
OSPAR COMPP and NCA differ from ASSETS inthe following aspects: the concentrations of DINand DIP are taken as indicators of eutrophication;and OSPAR COMPP fails to set thresholds forindicators concerned. This was initially intended toallow flexibility and discretion when applying it to arange of countries, but it also leads to ambiguity anduncertainty in the final results.
A general limitation of both the Chinese methodsand OSPAR COMPP and NCA is that there is nodifferential weighting across indicators. Althoughthe scientific community has been unable to agreeupon a single, reliable trophic state index, it iscommonly agreed that the relative importance ofindicators differs (Lu and Lo 2002; Bricker et al.2006). From the range of methods reviewed, onlyASSETS (and its NEEA predecessor) providesweighting of the various components, both withincomponents (Influencing Factors, Overall Eutro-phic Condition, and Future Outlook) and in theoverall aggregation procedure, where the FutureOutlook component weighs less in the overallclassification.
SELECTION OF STUDY SITES AND METHODS
The Changjiang Estuary and Jiaozhou Bay werechosen to compare the Chinese Nutrient IndexMethods and ASSETS, with respect to adequacy fora broad-scale assessment of Chinese coastal waters.The two systems have very different characteristics,as summarized in Table 5 (Editorial Board of Baysin China 1993, 1998). The Changjiang Estuary wasselected because it is the largest estuary in China (ofthe longest river in Asia), with the largest watershedand highest population density, and consequentlywith high nutrient loads that frequently lead toeutrophication problems. The ASSETS method hasbeen applied successfully to systems from differentecoregions representing all sizes in the U.S. andEurope (see http://www.eutro.org/syslist.aspx),from less than 1 km2 to about 7,000 km2, withvariable population densities and watershed uses.This includes the Mississippi-Atchafalaya RiverPlume (Rabalais et al. 2002), which has an areagreater than 12,000 km2 whose similarity to theChangjiang Estuary makes it appealing for aninternational comparison.
The interest in Jiaozhou Bay is mainly due to thetop-down control in the local ecosystem provided byshellfish aquaculture (Han and Wang 2001; Li et al.2005), which might suggest broader options for themanagement of eutrophication. As a typical medi-um-sized system (Table 1), it is a perfect test
TABLE 4. Summary of comparison among Phase I and Phase II methods (adapted from Bricker et al. 2006).
Methods Temporal Focus Indicator Criteria-Thresholds Combination Method
NutrientIndex I
Not specified Modified after Japanese criteria Sum of four ratios
NutrientIndex II
Not specified Modified after Japanese criteria Ratio of three indicators to their thresholdvalues
PCA Not specified Modified after Japanese criteria Comparisons among primary componentsand their threshold values
Fuzzy Analysis Not specified National standards Probabilities comparisonOSPAR COMPP Growing season, winter
for nutrientsIndividually/regionally determined
reference conditionIntegration of scores for four categories
EPA NCA Summer index periods Determined from American nationalstudies
Ratio of indicators: good/fair indicators topoor/missing data
ASSETS Annual cycle Determined from American nationalstudies
Average of primary and highest secondary arecombined by matrix
TABLE 5. Summary of two study sites.
Changjiang Estuary Jiaozhou Bay
Area (km2) 51,000 397Location East China (31u149N, 121u279E) East China (35u389–36u189N, 120u049–120u239E)Discharge (m3 yr21) 9.3 3 1011 8 3 108
Morphology Tubular estuary BayPressure Rural and industrial population 18 million
(Shanghai City)Rural and industrial population 8 million
(Qingdao City)Cultivated marine resources Kelp, shellfish ShellfishData availability Low HighProblems HABs, hypoxia HABs
Trophic Assessment in Chinese Coastal Systems 907
candidate prior to wider application in China. It iswell studied and the data availability lends confi-dence to the reliability of the assessment.
Chinese Nutrient Index Methods I and II are themajor assessment methods used in China and wereused here to evaluate the eutrophication status ofthe Changjiang Estuary and Jiaozhou Bay. ASSETSwas selected from the available Phase II approachesfor comparison tests, based on the followingfeatures. It has been successfully applied and testedfor 141 estuaries in the continental United States,10 estuaries in Portugal, and a number of coastalsystems in the U.K., Ireland, Italy, Germany, andAustralia (Bricker et al. 1999; Ferreira et al. 2003,2005; Bricker et al. 2007). It can accommodatediverse estuarine and coastal morphologies, riverdischarge, and tidal range conditions, together witha variety of system uses and environmental symp-toms. It was consolidated through intense peerreview within the scientific community, and hasbeen extensively published in the open literature(e.g., Bricker et al. 2003; Ferreira et al. 2005; Scaviaand Bricker 2006; Ferreira et al. 2007a; Whitall et al.2007). It takes direct and indirect eutrophicationsymptoms into account, when compared to NutrientIndex Methods and provides a more accurateevaluation than OSPAR COMPP (a full discussion
of comparisons among methods is given in thediscussion section).
CHANGJIANG ESTUARY
The 1.94 3 106 km2 Changjiang river basin ischaracterized by intense industrial and urbanactivity, especially in the lower reaches and estua-rine portion of the river (Editorial Board of Bays inChina 1998). It has a temperate climate and isheavily populated with an estimated population of400 million. The river discharge of 29,000 m3 s21
delivers about 480 million tons of sediment eachyear to the estuarine and coastal area. TheChangjiang River is a major source of nutrients tothe coastal zone and acts as a conduit thattransports anthropogenic loading from the catch-ment to the estuary and adjacent coastal waters(Chen and Chen 2003a). Located on a mesotidalcoast, the estuary is a wide, shallow, and partiallymixed system (Fig. 2).
The main issues of concern are HABs andhypoxia. HABs are frequently observed in theChangjiang estuary and extended coastal watersand are the primary environmental issue. The EastChina Sea is the area where the most severe HABsoccur among the four seas of China, accounting for36% of the total recorded number of blooms. Thefrequency of HAB occurrences as well as theduration and spatial extent of affected areas haveincreased significantly and continually since the1990s; in 2002, there were 51 individual HABoccurrences observed in the Changjiang estuaryand adjacent coastal areas (Guan and Zhan 2003).Toxic HAB genera, such as Alexandrium andGymnodrium, are often observed, resulting in killsof fish and zoobenthos, and have damaged nearbyfishing grounds such as the Zhoushan fishing area.
A secondary issue of concern in this area is theoccurrence of hypoxia in near-bottom waters off theChangjiang estuary and adjacent coastal waters,which has increased continuously since first record-ed in the 1950s (Li and Daler 2004; Fig. 2). Theconditions in the Changjiang and adjacent watersare very similar in scale to the low DO zone in theMississippi-Atchafalaya River Plume, which hasincreased in size since 1985, reaching the largestarea ever recorded in 2002 (Rabalais et al. 2002;Bricker et al. 2007).
Estimation of Nutrient Input to the Changjiang
Although ASSETS usually determines nutrientloading based on river discharge and concentra-tions of relevant nitrogen and phosphorus species,the Changjiang basin was treated differently, be-cause of uncertainties with respect to measured dataand to permit evaluation of catchment management
Fig. 2. Location map of Changjiang Estuary and estimatedhypoxic areas (adapted after Chen and Zhong 1999; Li etal. 2002).
908 Y. Xiao et al.
scenarios. Different watershed modeling approach-es were tested, as detailed below.
The first step was the collection of relevanthydrological and agricultural data on the Chang-jiang catchment, including: a topographic map,built using the GTOPO30 (Global Digital ElevationModel with a horizontal grid spacing of 30 arcseconds) data set with a resolution of 1 km2; and aland cover map with 1-km2 resolution, based on theU.S. Geological Survey’s Global Land Cover Char-acteristics Data Base. The topographic data set wasused to divide the Changjiang river basin into 87subbasins using an automatic watershed delineationmethod (the D8 method; Grayson and Bloschl2001), in order to further detail the spatial sourcesof nutrients within the catchment.
The Soil and Water Assessment Tool (SWAT) wasinitially applied to simulate the nutrient load intothe Estuary. SWAT is a physically-based model withthe objective of assessing the effect of landmanagement on water, sediment, and agriculturalpollution (for a full description see Neitsch et al.2002). SWAT proved to be an inappropriate tool tosimulate such a large area as the Changjiang riverbasin. Several authors have shown that the results ofhydrological models are significantly affected byproblems related to coarse-scale representation ofindicators and small-scale processes over large areas(e.g., Fisher et al. 1997; Bashford et al. 2002;Muttiah and Wurbs 2002; Venohr et al. 2005). Booij(2003) determined that the calculation of hydro-logical indicators related to topography should usedata sets with a minimum resolution of 100 3100 m, so the digital elevation map used in thisstudy (1,000 3 1,000 m) is inappropriate for thispurpose. Booij (2003) also determined that a basinshould be subdivided into modeling units of100 km2 or less, while the huge size of theChangjiang basin led to modeling units averaging11,600 km2. It appears that issues of scale preventthe application of a dynamic modeling tool to theChangjiang catchment; in spite of this, SWAT isrecognized as a useful tool for water qualitymanagement in smaller catchments (e.g., Arheimerand Olsson 2003; Santhi et al. 2006; Chaplot 2007)and can potentially be applied to the smallercatchments of other Chinese systems.
In the light of these results, an Export CoefficientModel (ECM) was chosen to estimate nutrientloading from the Changjiang catchment to theestuary, based on the watershed delineation de-tailed previously. The ECM is not dependent onhydrological process modeling, using instead landcover data maps to integrate the total annual basinnutrient loads from the many unique watershedareas, and then adding other nutrient sources suchas septic systems, wastewater treatment plants, and
precipitation (Reckhow and Simpson 1980). Thenutrient load in a river basin is obtained throughthe following equation:
LN ~XM
i~1
Ei | Ai½ �z S z W z P ð3Þ
where LN is the basin nutrient load (kg yr21); Ei isthe export coefficient (kg ha21 yr21) for land class i,Ai is the area of the watershed in land class i (ha), Sis the septic load (kg yr21), W is the wastewater load(kg yr21), and P is the precipitation load (kg yr21).
As a scoping model for estimating lumped annualbasin nutrient loads (Reckhow et al. 1980; Mattiklliaand Richards 1996; Johnes and Heathwaite 1997;Endreny and Wood 2003), ECM is a robust methodapplicable across many different watersheds (Beau-lac and Reckhow 1982; Clesceri et al. 1986; Frink1991; Line et al. 2002).
Unlike SWAT, ECM does not use meteorologicaldata or mechanistic pollutant-atmosphere-vegetation-soil equations, nor does it consider chemical process-es among nutrient species. Its modeling strength andadaptability have at least two advantages (Endrenyand Wood 2003): it is functional within watershedsthat meet the minimum data needs, and it remainssimple to use (Worrall and Burt 1999).
The areas used for ECM were obtained from thesubbasin delineation, while the nutrient coefficientswere collected from the literature (Reckhow et al.1980; Johnes 1996; Worrall and Burt 1999; Bernal etal. 2003). The annual average nutrient loads havebeen estimated as 11.4 ton N ha21 and 3.5 ton Pha21, with the highest specific export rates fromareas with intensive agriculture (double crop rice-lands and wheat + corn croplands). ECM coeffi-cients can be difficult to transfer among catchments(Wade et al. 2005), but they have been shown to beconsistent across broad land-use types (e.g., crop-land, pasture) for different watersheds (Harmel etal. 2006); Wade et al. (2005) report a broaddependency of river N concentration on catchmentland-use typology, rather than on specific agricul-tural practices, which supports the usefulness ofECM in determining the magnitude of N and Pexports from a given catchment (Wade et al. 2004).
JIAOZHOU BAY
Jiaozhou Bay (Fig. 4) is located on the west coastof the Yellow Sea (35u579–36u189N, 120u069–120u219E) with a surface area of 397 km2 andaverage depth of 7 m (Editorial Board of Bays inChina 1993). Jiaozhou Bay is a semi-enclosed waterbody, connected to the Yellow Sea through a 2.5-kmchannel, and has a mean tidal range of 2.5–3.0 m.The tides, which at spring tide can reach 4.2 m,induce strong turbulent mixing, resulting in nearly
Trophic Assessment in Chinese Coastal Systems 909
homogeneous vertical profiles of temperature andsalinity (Liu et al. 2004).
The bottom of Jiaozhou Bay contains spawning,nursery, and feeding grounds for fish. Over the lasttwo decades, intensive mariculture has been devel-oped. Historically, this has focused on the bayscallop (Argopecten irradians) and Pacific oyster(Crassostrea gigas), cultivated on longlines. Morerecently, the longlines have been removed and thesystem is currently used for cultivation of Manilaclam (Tapes philippinarum), with an estimatedproduction of 200,000 tons (total fresh weight)per year. Shellfish biomasses of this scale will filterthe entire bay in under one week, considering anaverage clearance rate of 1 l ind21 h21 for T.philippinarum (Hawkins personal communication).Top-down control from aquaculture has a poten-tially significant effect in reducing the expression ofeutrophication symptoms (Zhou et al. 2006).
The main issue in Jiaozhou Bay is the increase ofHABs, as both the frequency and magnitude of theHAB incidents have increased since 1990s, althoughmost events are non-toxic (Han et al. 2004). Themain HAB species include Biddulphia aurita, Eu-campia zoodiacus, Mesodinium rubrum, Noctiluca scin-tillans, and Skeletonema costatum (Wang et al. 2006).
Results and Discussion
Tables 6 and 7 present the results from theapplication of different eutrophication assessmentmethods for the Changjiang Estuary and forJiaozhou Bay.
CHANGJIANG ESTUARY
Influencing Factors
Nutrient Input. The results from the application ofECM indicate that the total N input into theChangjiang Estuary is 2.21 3 106 ton yr21 and theP input is 0.69 3 106 ton yr21. Figure 3 presents thedetails of nutrient distribution in the Changjiangriver basin. This N load, combined with systemvolume and mean salinity, indicates a High categoryfor the nutrient component of the InfluencingFactor score (0.9998).
Susceptibility. The susceptibility of the Changjiangis considered Moderate based on dilution andflushing capabilities. The dilution volume in theChangjiang Estuary was estimated as 6.4 3 1011 m3,with a mean thickness of 12.5 m in the upper layer.Mean salinities in this layer and offshore are 25 and30 psu, respectively, giving a dilution potential ofModerate. The flushing potential is consideredModerate, given a tidal range of 2.7 m and dischargeof 925 3 1011 m3 yr21 from the Changjiang River(Che et al. 2003). The combination of High loadand Moderate susceptibility gives a final InfluencingFactor rating of Moderate High.
Overall Eutrophic Condition
Primary Symptoms Method. Chl a is the onlyindicator with available data for evaluating primarysymptoms. No information on macroalgae wasreported in the literature, which was classified as
TABLE 7. ASSETS application to Jiaozhou Bay.
Index Method Indicator Level of Expression Index Result ASSETS Score
InfluencingFactors
Susceptibility Dilution potential Moderate Low (due to intenseshellfish aquaculture)
High
Flushing potential ModerateNutrient inputs High
Overall EutrophicCondition
Primary SymptomsMethod
Chlorophyll a Low LowMacroalgae No problem
Secondary SymptomsMethod
Dissolved oxygen LowSAV loss LowNuisance and toxic blooms Low
Future Outlook Future nutrient pressure Decrease Improve low
TABLE 6. Summary of ASSETS application to the Changjiang Estuary.
Index Method Indicator Level of Expression Index Result ASSETS Score
InfluencingFactors
Susceptibility Dilution potential Moderate Moderate High BadFlushing potential Moderate
Nutrient inputs HighOverall Eutrophic
ConditionPrimary Symptoms
MethodChlorophyll a Moderate HighMacroalgae Unknown
Secondary SymptomsMethod
Dissolved oxygen ModerateSAV loss UnknownNuisance and toxic blooms High
Future Outlook Future nutrient pressure Increase Worsen High
910 Y. Xiao et al.
Unknown. Detailed chl a data are difficult to find,but for indicative purposes, the literature data areused to carry out a pilot test (Zhou et al. 2004). Themean values of maximum concentration fall into theASSETS medium category (5–20 mg l21), but bloomsoccur over a huge area. The symptom level of chl aconcentration is considered to be Moderate.
Secondary Symptoms Method. DO and HABs areanalyzed as secondary symptoms, but no informa-tion was found for loss or change of SAV. TheChangjiang Estuary has long exhibited problemswith low DO (Li et al. 2002). Over the last twodecades, minimum values of DO in the low oxygenregion of the Changjiang Estuary have decreasedfrom 2.85 to 1 mg l21. A 1999 survey of the estuaryrevealed a 13,700-km2 bottom water hypoxic zone(, 2 mg l21) with an average thickness of 20 m anda minimum value of 1 mg l21 (Li et al. 2002). Theplume of oxygen-depleted water extended in asoutheasterly direction to the 100-m isobath, alongthe bottom of the continental shelf of the EastChina Sea (Fig. 2). These observations clearly
indicate that the system is under severe biologicalstress (, 5 mg l21). The combination of observedhypoxia that occurs on an annual basis and the largearea over which it occurs (26.9% of the system area)results in an expression level of DO of Moderate.
Along with the frequent reports of HAB inci-dents, the duration of blooms can last for weeks tomonths; a S. costatum bloom was reported from May10 to 23, 2001. Considering the high frequency andlong duration of occurrence of HABs in the estuary,the level of nuisance and toxic blooms falls into theHigh category.
Since the secondary symptom level is determinedby the highest value of three symptoms in ASSETS,the Changjiang Estuary is considered to fall into theHigh category. The overall eutrophic condition inthis system is considered to be High due toModerate primary symptoms and High secondarysymptoms.
Future Outlook
Based on China’s strategic planning for develop-ment, the Changjiang drainage basin is expected to
Fig. 3. Nutrient load in Changjiang river basin.
Trophic Assessment in Chinese Coastal Systems 911
provide an estimated 107–108 ton yr21 of additionalfood in order to feed the increasing populationwithin the next 50 yr. This will probably result in anexpansion of croplands, coupled with a furtherincrease in fertilizer application in a denselypopulated area that is already characterized byintensive agriculture. If DIN concentrations contin-ue to increase at the same rate as in the last twodecades, the DIN load will be an estimated 4.1 3106 ton yr21, twice as much as in 1998 (Zhang et al.1999b). One additional concern is that with theconstruction of the Three Gorges Dam, upstreamsilicate (Si) discharge is expected to decreasedrastically, leading to further decreases in theSi:N ratio and potentially introducing substantialchanges in phytoplankton composition. In short,the eutrophic status in Changjiang Estuary isexpected to worsen over the next few years. Table 6summarizes the results obtained for the applicationof ASSETS to Changjiang estuary, with an overallscore of Bad, indicating a high level of eutrophica-tion.
JIAOZHOU BAY
Influencing Factors
The volume of Jiaozhou Bay is 1,900 3 106 m3 andthe N load into the bay is 30 ton d21 (Wang et al.2006), which result in a High rating for the nutrientcomponent of Influencing Factor (0.933).
Strong tidal mixing and high river discharge (8 3108 m3 yr21) contribute to moderate flushing anddilution potential (Editorial Board of Bays in China1993). The intensive top-down control of thefoodweb has a significant effect on mitigatingeutrophic symptoms. In the 1960s, there was somekelp culture along the east coast of Jiaozhou. Sincethe 1980s, shrimp and shellfish culture have beendeveloping in the bay. In the 1990s, shellfish culturehas become more dominant (Fig. 4).
The susceptibility component of the InfluencingFactor based on only natural circumstances isconsidered Moderate but when shellfish aquacul-ture is taken into account, the overall susceptibilityis considered to be Low. This is one example of thedifficulty in universal application of this kind ofmethod, since ASSETS must be potentially adaptedto incorporate local societal factors. Another is thehuman consumption of Enteromorpha sp. in China,often produced in areas highly affected by sewagedischarge. The excessive growth of opportunisticmacroalgae, considered a primary eutrophicationsymptom in ASSETS, is not seen as a liability inmany parts of China.
The combination of High nutrient load and Lowsusceptibility gives an overall Influencing Factorrating of Moderate Low.
Overall Eutrophic Conditions
Primary Symptoms Method. Chl a is the onlyindicator with information for the primary symp-toms. No information was found for macroalgae,which was therefore classified as Unknown. Maxi-mum chl a values in Jiaozhou Bay did not exceed thethreshold indicated in ASSETS for Medium eutro-phic conditions. ASSETS uses the 90th percentilevalue to alleviate the extreme value problem, inorder to provide a more robust maximum concen-tration for chl a. In the bay this value is between 4–5 mg l21, i.e., in the Low category (Fig. 5). The ratingfor primary symptoms is Low based on chl a.
Secondary Symptoms Method. Discrete data for DOwere collected from various sites to cover one annualcycle. No information was found for SAV, butconsidering the large scale of kelp aquaculture in thebay, the level for this symptom would be at worst Low.
Very few values below the ASSETS threshold forbiologically stressful DO conditions (5 mg l21) weredetected in Jiaozhou Bay. As described earlier, the10th percentile value is applied to provide a moreconsistent minimum value for DO. In this system,the 10th percentile for annual DO data is between6–7 mg l21, indicating no problems for this indica-tor (Fig. 5).
Although HAB reports are not unusual, most ofthese are non-toxic (Han et al. 2004). According toHan et al., there were up to 69 harmful algal speciesobserved in Jiaozhou Bay. Toxic blooms areregistered episodically, and usually last only a fewdays; a S. costatum bloom was reported to last for fivedays in July 1998 (Huo et al. 2001). The symptom ofnuisance and toxic blooms is rated as Low.
In synthesis, the highest level of the threesecondary symptoms falls into the Low category,and the Overall Eutrophic Condition resulting from
Fig. 4. Location map of Jiaozhou Bay and the shellfish culturedistribution in late 1990s (adapted from Shen et al. 2006).
912 Y. Xiao et al.
the combination of primary and secondary symp-toms for this system is Low.
Future Outlook
The estimate based on the current developmentscenario gives a 9.3% population increase over 20 yr(P.R.C. National Bureau of Statistics 2001). Qingdao(the main land nutrient source, population of8 million) is strongly promoting its tourism industryand less space is available for mariculture inJiaozhou Bay. The reduced top-down control onprimary production could lead to increased eutro-phic symptoms.
Qingdao also prepares to host the OlympicSailing Regattas in 2008, which has focused atten-tion on water quality issues and mitigation ofeutrophic symptoms. The government has pledgedto build more wastewater treatment plants in thenear future and more restrictive pollutant emissionregulations are coming into effect (Wang et al.2006).
As a whole, nutrient loads are expected todecrease in spite of the increase in the urbanpopulation, and the water quality in Jiaozhou Bay islikely to improve. The Future Outlook can be
considered to be Improve Low. Table 7 summarizesthe results obtained from the application of ASSETSto Jiaozhou Bay, which resulted in an overall scoreof High, indicative of a system without eutrophica-tion problems.
Conclusions
COMPARISON OF RESULTS
The system classifications from the application ofthe ASSETS assessment method to ChangjiangEstuary and Jiaozhou Bay are Bad and High,respectively (Tables 6 and 7). While the assessmentfor the Changjiang is expected, the result forJiaozhou is better than expected as a result of top-down control related to intensive mariculture in thissystem. This has important implications for success-ful management of nutrient-related problems (seethe next section).
Comparison of the ASSETS classifications tothose from the application of the Chinese methodsshows differences among the results (Table 8).While the Chinese Nutrient Index Phase I methodsmay discriminate between problem and non-prob-lem areas, they are unable to indicate the degree ofseverity. ASSETS provides a more detailed classifi-cation of system eutrophication status. This be-comes more important when required managementmeasures are implemented and the success of themeasures must be examined. In the case of ASSETS,it is possible to measure incremental changes as asystem evolves, and management measures may bemodified as necessary. Smaller changes would notbe detected by the Chinese Nutrient Index Methodsand timely adaptive management would not bepossible.
The Nutrient Index Methods cannot by definitionbe clear indicators for a large system, because thereis no accommodation in the methods for spatialdifferences in level of effect within a water body.Nutrient Index Method II could not be successfullyapplied in Jiaozhou Bay because the large variabilityin results from different sampling sites made itimpossible to calculate an overall value. Theevaluation of systems by salinity zone, as in ASSETS,contributes to a more accurate evaluation of thesystem and subsystems, which is necessary to targetmanagement efforts.
Comparison of results for the Changjiang toresults from the application of ASSETS to the
Fig. 5. Frequency distributions for chlorophyll a and dissolvedoxygen in Jiaozhou Bay.
TABLE 8. Comparisons of results from different assessment methods.
Nutrient Index Method I Nutrient Index Method II ASSETS
Changjiang Estuary Eutrophic Eutrophic BadJiaozhou Bay Eutrophic Could not be applied High
Trophic Assessment in Chinese Coastal Systems 913
Mississippi-Atchafalaya River Plume show similarresults. The Mississippi-Atchafalaya River watersheddrains about 40% of the continental United States, acatchment area characterized by intensive agricul-ture. The overall eutrophic condition is High as aresult of high levels of chl a and of its mostcharacteristic problem, persistent hypoxia (Brickeret al. 2007). The Future Outlook is for worseningconditions due to expected population growth andincreased fertilizer use to grow food to supportincreased dietary needs (Turner et al. 2006). In2001, an action plan for reducing, mitigating, andcontrolling hypoxia in the Mississippi-AtchafalayaRiver Plume (Mississippi River/Gulf of MexicoWatershed Nutrient Task Force 2001) was en-dorsed by federal agencies, states, and tribalgovernments. The goal of reducing the hypoxiczone to an area , 5,000 km2 by the year 2015requires a N load reduction of 30–45%. Implemen-tation will be based on a series of voluntary andincentive-based activities, including proper timingand amount of fertilizer applications, best manage-ment practices on agricultural lands, wetlandrestoration and creation, river hydrology remedia-tion, riparian buffer strips, and nutrient removalfrom storm water and wastewater. Comparison ofthe results of the success of the Mississippi-Atch-afalaya River management plan with plans forreduction of the Changjiang hypoxic zone mayhelp to promote more successful management inboth systems.
FUTURE DEVELOPMENTS AND WIDER APPLICATION
The ECM results show that most of the N and Ploads from the Changjiang river basin come fromthe eastern part of the catchment as a result ofintensive agricultural activity. This informationshould be used to provide the basis for targetedmanagement. As an example, focusing more inten-sive management efforts on agriculture in theeastern areas of the basin will provide greatersuccess in reducing nutrient-related problems thanif management efforts were applied with equalintensity across the basin.
The top-down control of the foodweb in JiaozhouBay suggests a feasible way to manage the eutrophi-cation in a coastal system. These strategies foreutrophication control, which have traditionallybeen used in China for many years, are beingdiscussed with respect to practical implementationin the EU and the U.S. (e.g., Lindahl et al. 2005;Ferreira et al. 2007b). Paradoxically, the Chinese,U.S., and other governments and scientists currentlyfocus mainly on a bottom-up approach in improvingwater quality, though there is still plenty of scope topromote top-down control in the food chain. Waterquality data from an annual program with monthly
measurements at seven stations in Jiaozhou Bayduring 1999–2000 were used to estimate the grossremoval of algae by Manila clams. On the basis ofreported bivalve stocks, these organisms annuallyremove about 627 ton yr21 of chl a, which (consid-ering a carbon:chlorophyll ratio of 50 and thestandard Redfield C:N ratio of 45:7 in mass)corresponds to the removal of almost 4,900 ton yr21
of N. This equates to the annual discharge of about1.5 million people, or 17% of the population ofQingdao, and to about 45% of the estimated11,000 ton yr21 N load. Along with economicbenefits, the introduction of filter feeders on areasonable scale allows for cost-effective removalof nutrients and mitigation of eutrophic con-ditions, which is more environmentally friendlyand sustainable for a coastal system (Shastri andDiwekar 2006).
Environmental managers should be cautiouswhen reducing mariculture, since doing so couldlead to worsening eutrophic conditions. There is aproposal in Qingdao city to limit aquaculture inJiaozhou Bay as part of future management plans,partly driven by the Olympic sailing regattas in 2008.As mentioned previously, shellfish aquacultureappears to be the major reason why this bay is notseverely eutrophic given the high nutrient load.Decrease of shellfish culture may result in moresevere eutrophic symptoms in Jiaozhou Bay.
Compared to Chinese assessment methods,ASSETS proved to be a more feasible method toapply to coastal systems based on these two casestudies. Data acquisition proved to be the mostdifficult part of this study, and it is expected thatdata limitations will be a challenge in a morecomprehensive assessment of Chinese coastal sys-tems. It is recommended that additional data becollected for the two study sites and that monitoringof other sites should also be a priority, to providethe basis for a national coastal eutrophicationassessment. In some cases, particularly for subtrop-ical systems, it may be necessary to make adjust-ments to thresholds for indicators such as chl a. Inthe application of ASSETS in the U.S., the rangesand thresholds determined by the group of eutro-phication experts accurately reflected conditions inall 141 systems, with the exception of Florida Bay, asensitive system in the Gulf of Mexico region. Forthis system, a lower threshold was establishedheuristically.
There is a clear need to better understand thecurrent eutrophication status of Chinese coastalwater bodies, given concerns about eutrophication-related degradation during the past two decadesand perspectives for an increase in nutrient-relatedpressures on the coastal zone over the coming years.The successful assessment of these test systems,
914 Y. Xiao et al.
despite the differences in size, physical characteris-tics, and uses, suggests that the ASSETS methodmay potentially be recommended for a Chinesenational eutrophication assessment. Such an assess-ment will provide the necessary support to coastalmanagers, by providing an overview of the scale ofthe problems, highlighting the areas where prioritymanagement plans should be developed, andinforming integrated coastal zone management,bearing in mind the critical effects that changes inbivalve mariculture may have on eutrophicationsymptoms.
Many of the aspects discussed herein are ofpotential interest in other parts of southeast Asia,which share with China the challenges of balancingrapid economic growth with improvement ofenvironmental quality in the coastal zone. In thiscontext, rural and peri-urban activities such asintegrated multitrophic aquaculture have overthousands of years provided natural remediationfor the excessive discharge of nutrients to bays andestuaries.
ACKNOWLEDGMENTS
The authors would like to thank the Erasmus Mundusprogramme, the INCO-CT-2004-510706 SPEAR Leverage Pro-gramme - TAICHI project, and the 2002CB412406 project of theChina Key Basic Research Programme for supporting this work.We are grateful to A. Newton, Erasmus Mundus Water and CoastalManagement M.Sc. Coordinator, T. Simas for advice on method-ological issues, and N. Pacheco for support in SWAT and GISapplications. The contributions of two anonymous reviewers aregratefully acknowledged.
LITERATURE CITED
ARHEIMER, B. AND J. OLSSON. 2003. Integration and Coupling ofHydrological Models with Water Quality Models: Applicationsin Europe. RA VI - Europe working group on hydrology report,World Meteorological Organization, Geneva, Switzerland
BASHFORD, K. E., K. J. BEVEN, AND P. C. YOUNG. 2002. Observationaldata and scale-dependent parameterizations: Explorationsusing a virtual hydrological reality. Hydrological Processes 16:293–312.
BEAULAC, M. N. AND K. H. RECKHOW. 1982. An examination of landuse-nutrient export relationships. Water Resources Bulletin 18:1013–1024.
BERNAL, S., A. BUTTURINI, E. NIN, F. SABATER, AND S. SABATER. 2003.Leaf litter dynamics and nitrous oxide emission in a Mediter-ranean riparian forest: Implications for soil nitrogen dynamics.Journal of Environmental Quality 32:191–197.
BOOIJ, M. J. 2003. Determination and integration of appropriatespatial scales for river basin modelling. Hydrological Processes 17:2581–2598.
BRICKER, S. B., C. G. CLEMENT, D. E. PIRHALLA, S. P. ORLANDO, AND
D. R. G. FARROW. 1999. National Estuarine Eutrophi-cation Assessment. Effects of Nutrient Enrichment in theNation’s Estuaries. National Oceanic and Atmospheric Admin-istration, National Ocean Service, Special Projects Office andNational Centers for Coastal Ocean Science, Silver Spring,Maryland.
BRICKER, S. B., J. G. FERREIRA, AND T. SIMAS. 2003. An integratedmethodology for assessment of estuarine trophic status.Ecological Modelling 169:39–60.
BRICKER, S. B., D. LIPTON, A. MASON, M. DIONNE, D. KEELEY, C.KRAHFORST, J. LATIMER, AND J. PENNOCK. 2006. ImprovingMethods and Indicators for Evaluating Coastal Water Eutro-phication: A Pilot Study in the Gulf of Maine. National Centersfor Coastal Ocean Science, Center for Coastal Monitoring andAssessment, Silver Spring, Maryland.
BRICKER, S. B., B. LONGSTAFF, W. DENNISON, A. JONES, J. WOERNER,C. WICKS, AND K. BOICOURT. 2007. National Estuarine Eutrophi-cation Assessment: Effects of Nutrient Enrichment in theNation’s Estuaries 1999–2004. National Oceanic and Atmo-spheric Administration Coastal Ocean Program DecisionAnalysis Series No. 26. National Centers for Coastal OceanScience, Silver Spring, Maryland.
BURKHOLDER, J. M., M. A. MALLIN, AND H. B. GLASGOW, JR. 1999.Fish kills, bottom water hypoxia and the toxic Pfiesteria complexin the Neuse River and Estuary. Marine Ecological Progress Series179:301–310.
CARLSON, R. 1977. A trophic state index for lakes. Limnology andOceanography 22:361–369.
CHAI, C., Z. YU, X. SONG, AND X. CAO. 2006. The status andcharacteristics of eutrophication in the Yangtze River (Chang-jiang) Estuary and the adjacent East China Sea, China.Hydrobiologia 563:313–328.
CHANG, N. B., H. W. CHEN, AND S. K. NING. 2001. Identification ofriver water quality using the Fuzzy Synthetic Evaluationapproach. Journal of Environmental Management 63:293–305.
CHAPLOT, V. 2007. Water and soil resources response to risinglevels of atmospheric CO2 concentration and to changes inprecipitation and air temperature. Journal of Hydrology 337:159–171.
CHE, Y., Q. HE, AND W. Q. LIN. 2003. The distributions ofparticulate heavy metals and its indication to the transfer ofsediments in the Changjiang Estuary and Hangzhou Bay,China. Marine Pollution Bulletin 46:123–131.
CHEN, J. Y. AND S. L. CHEN. 2003a. Ecological environmentalchanges in the Changjiang estuary and suggestions forcountermeasure. Water Resources Hydropower Engineering 34:19–25.
CHEN, Q. AND A. E. MYNETT. 2003b. Integration of datamining techniques and heuristic knowledge in fuzzy logicmodelling of eutrophication in Taihu Lake. Ecological Modelling162:55–67.
CHEN, B., J. WANG, J. TANG, AND S. WEN. 2002. Prediction to trendof nutrient status in Meizhou Bay, Fujian. Journal of Oceanogra-phy in Taiwan Strait 21:322–327.
CHEN, X. AND Y. ZHONG. 1999. Major impacts of sea-level rise onagriculture in the Yangtze delta area around Shanghai. AppliedGeography 19:69–84.
CLESCERI, N. L., S. J. CURRAN, AND R. I. SEDLAK. 1986. Nutrientloads to Wisconsin lakes: Part I. Nitrogen and phosphorusexport coefficients. Water Resources Bulletin 22:983–989.
CLOERN, J. E. 2001. Our evolving conceptual model of the coastaleutrophication problem. Marine Ecology Progress Series 210:223–253.
DETTMANN, E. H. 2001. Effect of water residence time on annualexport and denitrification of nitrogen in estuaries: A modelanalysis. Estuaries 24:481–490.
EDITORIAL BOARD OF BAYS IN CHINA. 1993. Bays in China. OceanPress, Beijing, China.
EDITORIAL BOARD OF BAYS IN CHINA. 1998. Bays in China. OceanPress, Beijing, China.
ENDRENY, T. A. AND E. F. WOOD. 2003. Watershed weighting ofexport coefficients to map critical phosphorous loading areas.Journal of the American Water Resources Association 39:165–181.
EUROPEAN COMMUNITY. 2000. Directive of the European Parlia-ment and of the Council 2000/60/EC, establishing a Frame-work for Community Action in the Field of Water Policy.
FERREIRA, J. G., S. B. BRICKER, AND T. C. SIMAS. 2007a. Applicationand sensitivity testing of a eutrophication assessment method
Trophic Assessment in Chinese Coastal Systems 915
on coastal systems in the United States and European Union.Journal of Environmental Management 82:433–445.
FERREIRA, J. G., A. J. S. HAWKINS, AND S. B. BRICKER. 2007b.Management of productivity, environmental effects and profit-ability of shellfish aquaculture – the Farm AquacultureResource Management (FARM) model. Aquaculture 264:160–174.
FERREIRA, J. G., T. SIMAS, A. NOBRE, M. C. SILVA, K. SCHIFFEREGGER,AND J. LENCART-SILVA. 2003. Identification of Sensitive Areas andVulnerable Zones in Transitional and Coastal PortugueseSystems. Application of the United States National EstuarineEutrophication Assessment to the Minho, Lima, Douro, Ria deAveiro, Mondego, Tagus, Sado, Mira, Ria Formosa: andGuadiana systems, IMAR Institute of Marine Research, Lisbon,Portugal.
FERREIRA, J. G., W. J. WOLFF, T. C. SIMAS, AND S. B. BRICKER. 2005.Does biodiversity of estuarine phytoplankton depend onhydrology? Ecological Modelling 187:513–523.
FISHER, P., R. J. ABRAHART, AND W. HERBINGER. 1997. The sensitivityof two distributed non-point source pollution models to thespatial arrangement of the landscape. Hydrological Processes 11:241–252.
FRINK, C. R. 1991. Estimating nutrient exports to estuaries. Journalof Environmental Quality 20:717–724.
GRAYSON, R. AND G. BLOSCHL. 2001. Spatial Patterns in CatchmentHydrology - Observations and Modelling. Cambridge UniversityPress, Cambridge, Massachusetts.
GUAN, D. M. AND X. W. ZHAN. 2003. Red tide disaster in coastalwater of China and its prevention suggestions. Marine Environ-mental Science 22:60–63.
GUO, W., X. ZHANG, Y. YANG, AND M. HU. 1998. Potentialeutrophication assessment for Chinese coastal waters. Journalof Oceanography in Taiwan Strait 17:64–70.
HAN, X. F. AND R. WANG. 2001. The grazing impact and regulationeffects of zooplankton on phytoplankton bloom. Marine Sciences(in Chinese) 25:31–33.
HAN, X. T., J. Z. ZOU, AND Y. S. ZHANG. 2004. Harmful algae bloomspecies in Jiaozhou Bay and the features of distribution. MarineSciences (in Chinese) 28:49–54.
HAO, J., W. HUO, AND Z. YU. 2000. Preliminary study on redtide occurrence in relation to nutritional condition inaquaculture seawater of Jiaozhou Bay. Marine Sciences (inChinese) 24:37–41.
HARMEL, D., S. POTTER, P. CASEBOLT, K. RECKHOW, C. GREEN, AND
R. HANEY. 2006. Compilation of measured nutrient load datafor agricultural land uses in the United States. Journal of theAmerican Water Resources Association 42:1163–1178.
HARRISON, P. J., M. H. HU, Y. P. YANG, AND X. LU. 1990. Phosphatelimitation in estuarine and coastal waters of China. Journal ofExperimental Marine Biology Ecology 140:79–87.
HAUXWELL, J., J. CEBRIAN, AND I. VALIELA. 2003. Eelgrass (Zosteramarina) loss in temperate estuaries: Relationships to land-derived nitrogen loads and effect of light limitation imposed byalgae. Marine Ecological Progress Series 247:59–73.
HUANG, X. P., L. M. HUANG, AND W. Z. YUE. 2003. Thecharacteristics of nutrients and eutrophication in thePearl River estuary, South China. Marine Pollution Bulletin 47:30–36.
HUO, W. Y., Z. M. YU, J. Z. ZOU, X. X. SONG, AND J. H. HAO. 2001.Outbreak of Skeletonema costatum red tide and its relations toenvironmental factors in Jiaozhou Bay. Ocenologia et LimnologiaSinica 32:311–318.
JOHNES, P. J. 1996. Evaluation and management of the impact ofland use on the nitrogen and phosphorus load delivered tosurface waters: The export coefficient modelling approach.Journal of Hydrology 183:323–349.
JOHNES, P. J. AND A. L. HEATHWAITE. 1997. Modeling the impact ofland use change on water quality in agricultural catchments.Hydrological Processes 11:269–286.
KOMPARE, B., I. BRATKO, F. STEINMAN, AND S. DZEROSKI. 1994. Usingmachine learning techniques in the construction of models I.Introduction. Ecological Modelling 75:617–628.
KUNG, H., L. YING, AND Y. C. LIU. 1992. A complementary tool towater quality index: Fuzzy clustering analysis. Water ResourcesBulletin 28:525–534.
LI, D. AND D. DALER. 2004. Ocean pollution from land-basedsources: East China Sea, China. Ambio 33:107–112.
LI, D., J. ZHANG, D. HUANG, Y. WU, AND J. LIANG. 2002. Oxygendepletion off the Changjiang (Yangtze River) Estuary. Science inChina 45:1137–1146.
LI, C. L., F. ZHANG, X. SHEN, B. YANG, Z. L. SHEN, AND S. SUN. 2005.Concentration, distribution and annual fluctuation of chloro-phyll-a in the Jiaozhou Bay. Oceanologia et Limnologia Sinica 36:499–506.
LIN, R. 1996. Review of assessing methods for coastal eutrophi-cation. Marine Environmental Science 15:28–31.
LIN, X., C. HUANG, F. LIN, AND X. CHEN. 2004. Principalcomponent-cluster analysis method for the assessment ofseawater eutrophication. Mathematics in Practice and Theory 34:69–74.
LINDAHL, O., R. HART, B. HERNROTH, S. KOLLBERG, L. LOO, L.OLROG, A. REHNSTAM-HOLM, J. SVENSSON, S. SVENSSON, AND U.SYVERSEN. 2005. Improving marine water quality by musselfarming: A profitable solution for Swedish society. Ambio 34:131–138.
LINE, D. E., N. M. WHITE, D. L. OSMOND, G. D. JENNINGS, AND C. B.MOJONNIER. 2002. Pollutant export from various land uses inthe Upper Neuse River Basin. Water Environment Research 74:100–108.
LIU, Z., H. WEI, G. LIU, AND J. ZHANG. 2004. Simulation of waterexchange in Jiaozhou Bay by average residence time approach.Estuarine, Coastal and Shelf Science 61:25–35.
LU, R. S. AND S. L. LO. 2002. Diagnosing reservoir water qualityusing self-organizing maps and fuzzy theory. Water Research 36:2265–2274.
MARCHINI, A. AND C. MARCHINI. 2006. A fuzzy logic modelto recognise ecological sectors in the lagoon of Venicebased on the benthic community. Ecological Modelling 193:105–118.
MATTIKALLI, N. M. AND K. S. RICHARDS. 1996. Estimation of surfacewater quality changes in response to land use change:Application of the export coefficient model using RemoteSensing and Geographic Information System. Journal ofEnvironmental Management 48:263–282.
MAY, C. L., J. R. KOSEFF, L. V. LUCAS, J. E. CLOERN, AND D. H.SCHOELLHAMER. 2003. Effects of spatial and temporal variabilityof turbidity on phytoplankton blooms. Marine Ecological ProgressSeries 254:111–128.
METTERNICHT, G. 2001. Assessing temporal and spatial changes ofsalinity using fuzzy logic, remote sensing and GIS. Foundationsof an expert system. Ecological Modelling 144:163–179.
MISSISSIPPI RIVER / GULF OF MEXICO WATERSHED NUTRIENT TASK
FORCE. 2001. Action Plan for Reducing, Mitigating, andControlling Hypoxia in the Northern Gulf of Mexico. Mis-sissippi River / Gulf of Mexico Watershed Nutrient Task Force,Washington, D.C.
MORIHIRO, A., A. OUTOSKI, T. KAWAI, M. HOSOME, AND K. MURAOKA.1981. Application of modified Carlson’s trophic state index toJapanese lakes and its relationship to other parameters relatedto trophic state. Research Report of National Institute of Environ-mental Studies 23:12–30.
MUTTIAH, R. S. AND R. A. WURBS. 2002. Scale-dependent soil andclimate variability effects on watershed water balance of theSWAT model. Hydrological Processes 256:264–285.
NEITSCH, S. L., J. C. ARNOLD, J. R. KINIRY, AND J. R. WILLIAMS. 2002.Soil and Water Assessment Tool User’s Manual: Version 2000.Blackland Research Center, Texas Agricultural ExperimentStation, Temple, Texas.
916 Y. Xiao et al.
NUNES, J. P., J. G. FERREIRA, F. GAZEAU, J. LENCART-SILVA, X. L.ZHANG, M. Y. ZHU, AND J. G. FANG. 2003. A model for sustainablemanagement of shellfish polyculture in coastal bays. Aquaculture219:257–277.
OKAICHI, T. 2004. Red Tides. Terra Scientific PublishingCompany, Tokyo, Japan.
OSPAR COMMISSION. 2003. OSPAR Integrated Report 2003 on theEutrophication Status of the OSPAR Maritime Area Based Uponthe First Application of the Comprehensive Procedure. OSPARCommission, London, U.K. http://www.ospar.org/documents/dbase/publications/p00189_Eutrophication%20Status%20Report%202003.pdf.
PARINET, B., A. LHOTE, AND B. LEGUBE. 2004. Principal componentanalysis: An appropriate tool for water quality evaluation andmanagement—application to a tropical lake system. EcologicalModelling 178:295–311.
PEI, H. AND J. MA. 2002. Study on the algal dynamic model forWest Lake, Hangzhou. Ecological Modelling 148:67–77.
PENG, Y. H. AND Z. D. WANG. 1991. Assessment of theeutrophication level in the estuary of Zhujiang River. MarineEnvironmental Science 10:7–13.
P.R.C. NATIONAL BUREAU OF STATISTICS. 2001. Major Figures on2000 Population Census of China. China Statistics Press,Beijing, China.
P.R.C. STATE OCEANIC ADMINISTRATION. 2007. The Official Gazettesof Environment Quality of China Sea in 2006. http://www.soa.gov.cn
RABALAIS, N. N., R. E. TURNER, AND D. SCAVIA. 2002. Beyondscience into policy: Gulf of Mexico hypoxia and the MississippiRiver. Bioscience 52:129–142.
RECKHOW, K. H., M. N. BEAULAC, AND J. T. SIMPSON. 1980.Modeling Phosphorus Loading and Lake Response underUncertainty: A Manual and Compilation of Export Coefficients.Michigan State University, East Lansing, Michigan.
RECKHOW, K. H. AND J. T. SIMPSON. 1980. A procedure usingmodelling and error analysis for the prediction of lakephosphorous concentration from land use information. Cana-dian Journal of Fisheries and Aquatic Science 37:1439–1448.
SALSKI, A. 1992. Fuzzy knowledge-based models in ecologicalresearch. Ecological Modelling 63:103–112.
SANTHI, C., R. SRINIVASAN, J. G. ARNOLD, AND J. R. WILLIAMS. 2006.A modeling approach to evaluate the impacts of water qualitymanagement plans implemented in a watershed in Texas.Environmental Modelling and Software 21:1131–1157.
SCAVIA, D. AND S. B. BRICKER. 2006. Coastal eutrophicationassessment in the United States. Biogeochemistry 79:187–208.
SCIRUS. 2007. Scientific Research Website. Amsterdam, TheNetherlands. http://www.scirus.com/
SHASTRI, Y. AND U. DIWEKAR. 2006. Sustainable ecosystemmanagement using optimal control theory: Part 1 (determin-istic systems). Journal of Theoretical Biology 241:506–521.
SHEN, Z. L. 2001. Historical changes in nutrient structure and itsinfluences on phytoplankton composition in Jiaozhou Bay.Estuarine, Coastal and Shelf Science 52:211–224.
SHEN, Z. L., Q. LIU, Y. L. WU, AND Y. YAO. 2006. Nutrient structureof seawater and ecological responses in Jiaozhou Bay, China.Estuarine, Coastal and Shelf Science 69:299–307.
STEVENSON, J. C., L. W. STAVER, AND K. W. STAVER. 1993. Waterquality associated with survival of submersed aquatic vegetationalong an estuarine gradient. Estuaries 16:346–361.
STRAIN, P. M. AND P. A. YEATS. 1999. The relationships bet-ween chemical measures and potential predictors of theeutrophication status of inlets. Marine Pollution Bulletin 38:1163–1170.
TETT, P., L. GILPIN, H. SVENDSEN, C. P. ERLANDSSON, U. LARSSON, S.KRATZER, E. FOUILLAND, C. JANZEN, J. LEE, C. GRENZ, A. NEWTON,J. G. FERREIRA, T. FERNANDES, AND S. SCORY. 2003. Eutrophica-tion and some European waters of restricted exchange.Continental Shelf Research 23:1635–1671.
TETT, P., R. GOWEN, D. MILLS, T. FERNANDES, L. GILPIN, M.HUXHAM, K. KENNINGTON, P. READ, M. SERVICE, M. WILKINSON,AND S. MALCOLM. 2007. Defining and detecting undesirabledisturbance in the context of marine eutrophication. MarinePollution Bulletin 55:282–297.
TOMASKO, D. A., D. L. BRISTOL, AND J. A. OTT. 2001. Assessment ofpresent and future nitrogen loads, water quality, and seagrass(Thalassia testudinum) depth distribution in Lemon Bay,Florida. Estuaries 24:926–938.
TURNER, R. E., N. N. RABALAIS, AND D. JUSTIC. 2006. Pre-dicting summer hypoxia in the northern Gulf of Mexico:Riverine N, P and Si loading. Marine Pollution Bulletin 52:139–148.
U.S. ENVIRONMENTAL PROTECTION AGENCY (USEPA). 2005. Nation-al Coastal Condition Report II. USEPA, Washington, D.C.
VENOHR, M., H. BEHRENDT, AND W. KLUGE. 2005. The effectsof different input data and their spatial resolution on theresults obtained from a conceptual nutrient emissionsmodel: The River Stor case study. Hydrological Processes 19:3501–3515.
VOLLENWEIDER, R. A. 1968. The scientific basis for lake and streameutrophication with particular reference to phosphorus andnitrogen as eutrophication factors. Technical Report DAS/DS1/68.27, Organization for Economic Co-operation andDevelopment, Paris, France.
VOLLENWEIDER, R. A. 1975. Input-output models with specialreference to the phosphorus loading concept in limnology.Schweizerische Zeitschrift fur Hydrologie 37:53–84.
WADE, A. J., C. NEAL, P. G. WHITEHEAD, AND N. J. FLYNN. 2005.Modelling nitrogen fluxes from the land to the coastal zone inEuropean systems: A perspective from the INCA project. Journalof Hydrology 304:413–429.
WADE, A. J., P. G. WHITEHEAD, H. P. JARVIE, C. NEAL, H. PRIOR, AND
P. J. JOHNES. 2004. Nutrient monitoring, simulation andmanagement within a major lowland UK river system: TheKennet. Mathematics and Computers in Simulation 64:307–317.
WANG, B. D. 2005b. Eutrophication assessment models forestuarine and coastal waters. Advances in Marine Science 23:82–86.
WANG, B. 2006. Cultural eutrophication in the Changjiang(Yangtze River) plume: History and perspective. Estuarine,Coastal and Shelf Science 69:471–477.
WANG, X. L., K. Q. LI, AND X. Y. SHI. 2006. The MarineEnvironmental Carrying Capacity of Major Pollutants inJiaozhou Bay. Science Press, Beijing, China.
WAZNIAK, C. E. AND P. M. GLIBERT. 2004. Potential impacts ofbrown tide, Aureococcus anophagefferens, on juvenile hard clams,Mercenaria mercenaria, in the coastal bays of Maryland, USA.Harmful Algae 3:321–329.
WHITALL, D., S. B. BRICKER, J. G. FERREIRA, A. M. NOBRE, T. SIMAS,AND M. C. SILVA. 2007. Assessment of Eutrophication inEstuaries: Pressure-State-Response and Nitrogen Source Appor-tionment. Environmental Management 40:678–690.
WORRALL, F. AND T. P. BURT. 1999. The impact of land-use changeon water quality at the catchment scale: The use of exportcoefficient and structural models. Journal of Hydrology 221:75–90.
XIONG, D. Q. AND S. Y. CHEN. 1993. Theoretical fuzzy model forthe eutrophication assessment in seawater. Marine Environmen-tal Science 12:104–110.
YAN, T., M. J. ZHOU, AND J. Z. ZOU. 2002. National Report of HABs inChina. www.pices.int/publications/scientific_reports/Report23/HAB_China.pdf
YAO, Y. AND Z. SHEN. 2005. A review on eutrophication research ofcoastal waters. Marine Sciences (in Chinese) 29:53–57.
ZHANG, J., Z. G. YU, J. T. WANG, J. L. REN, H. T. CHEN, H. XIONG,L. X. DONG, AND W. Y. XU. 1999a. The subtropical Zhujiang(Pearl River) estuary: Nutrient, trace species and theirrelationship to photosynthesis. Estuarine, Coastal and Shelf Science49:385–400.
Trophic Assessment in Chinese Coastal Systems 917
ZHANG, J., Z. F. ZHANG, S. M. LIU, Y. WU, H. XIONG, AND H. T.CHEN. 1999b. Human impacts on the large world rivers: Wouldthe Changjiang (Yangtze River) be an illustration? GlobalBiogeochemical Cycles 13:1099–1105.
ZHOU, Y., H. YANG, H. HU, Y. LIU, Y. MAO, H. ZHOU, X. XU, AND F.ZHANG. 2006. Bioremediation potential of the macroalgaGracilaria lemaneiformis (Rhodophyta) integrated into fed fishculture in coastal waters of north China. Aquaculture 252:264–276.
ZHOU, W., X. YUAN, W. HUO, AND K. YIN. 2004. Distribution ofchlorophyll a and primary productivity in the adjacent sea areaof Changjiang River Estuary. Acta Oceanologica Sinica 26:143–150.
ZOU, J., L. DONG, AND B. QIN. 1985. Preliminary studies oneutrophication and red tide problems in Bohai Bay. Hydro-biologia 127:27–30.
SOURCE OF UNPUBLISHED MATERIALS
HAWKINS, A. J. S. personal communication. Plymouth MarineLaboratory, The Hoe, Plymouth PL1 3DH, Devon, U.K.
Received, April 25, 2007Revised, July 10, 2007
Accepted, July 20, 2007
918 Y. Xiao et al.
Related Documents
